This project will maximize the metabolic information which can be obtained from localized in vivo 31P and 1H spectra by developing and applying new technical and data processing procedures. The sensitivity of the data will be improved by implementing automatic shimming, by installing proton decoupling for 31P spectroscopy and by extending water suppression techniques to use with 1H chemical shift imaging (CSI). The measurement of T1's and determination of absolute concentrations will also be important in accurately interpreting the data. In addition to these technical advances, new data processing procedures and software packages will be developed for examination in planning and automatic first pass analysis of the CSI data. Current methods for quantifying individual spectra and producing metabolic images will be extended and optimized. The information in coil sensitivity maps, relaxation times and peak areas of individual spectra will be combined to produce absolute metabolic concentrations, which will then be correlated with the anatomy as depicted in the associated proton images. To test the procedures and build up a database of metabolite concentrations in normal tissue, 31P CSI will be applied to study liver, muscle and brain of volunteers. In the case of brain, a comparison will also be made with the metabolic information provided by localized 1H spectroscopy. In the final years of the grant, the techniques will be applied to study two tumor systems: squamous cell carcinoma metastases of the neck and brain metastases. This will provide an opportunity to study tumor heterogeneity, to compare localized 31P and 1H spectra and to test the feasibility of applying our procedures in a clinical setting.